Don’t worry, be happy 😀

Face detection + Age/Gender Recognition + Emotion Recognition with Python & Tensorflow

Installation

Clone the repository.

Install these dependencies with pip3 install <module name>

• tensorflow
• numpy
• scipy
• imageio
• opencv-python
• pillow
• pandas
• matplotlib
• h5py
• keras

Compile your config.json inside config folder.

Now, you can run the project:
python3 app.py

or using pm2.

Models

In constants.py you can find the models that I used to detect faces, age, gender and emotions.
I didn’t include these models in my repository, but you can download them:

• HAAR Cascade Frontal Face
• Emotion Model (fer2013_mini_XCEPTION)
• Age/Gender Model

To train new models for emotion classification

• Download the fer2013.tar.gz file from here
• Move the downloaded file to the datasets directory inside this repository.
• Untar the file:
  tar -xzf fer2013.tar
• Download train_emotion_classifier.py from orriaga’s repo here
• Run the train_emotion_classification.py file:
  python3 train_emotion_classifier.py

Credits

• Computer vision powered by OpenCV.
• Neural network scaffolding powered by Keras with Tensorflow
• Convolutional Neural Network (CNN) deep learning architecture is from this research paper
• Pretrained Keras model and much of the OpenCV code provided by GitHub user oarriaga
• Emotion recognition based on petercunha/Emotion
• Age and gender recognition based on Tony607/Keras_age_gender